Summary The Messershmitt-Bolkow-Blohm (MBB) BO105helicopter (registration C-GGGC, serial numberS617) was being used for various tasks associated with the upkeep and operation of lighthouse and coastal navigation facilities in the Burin Peninsula area of Newfoundland and Labrador. While returning to Marystown in the late afternoon of 07December2005, with one pilot and one passenger on board, the helicopter encountered heavy snow showers and, at about 1628 Newfoundland standard time, the helicopter crashed into the water of MortierBay, east of Marystown. Both the pilot and the passenger survived the water impact and escaped from the helicopter. However, the pilot perished from hypothermia, and the passenger drowned. Ce rapport est galement disponible en franais. Other Factual Information Weather The Atlantic graphic area forecast (GFA)1 weather chart, issued at 0751Newfoundland standard time,2 forecast the weather in the BurinPeninsula area, Newfoundland and Labrador, to be ceilings of 2500feet above sea level (asl), visibility of two to six statute miles (sm) in light snow showers with scattered areas of stratocumulus clouds giving ceilings of 400feet asl and visibilities of sm in snow. The winds were forecast to be from the west gusting to 30knots. History of the Flight The Canadian Coast Guard (CCG) helicopter, operated by Transport Canada (TC) Aircraft Services Directorate (ASD) as call signCG352, was based in St.John's, Newfoundland and Labrador. On the day of the accident, the first task was to move personnel and supplies to the Green Island lighthouse, which is located 7nautical miles (nm) off the southern tip of the BurinPeninsula. CG352 was then to proceed to Marystown, Newfoundland and Labrador, to pick up a CCG technician and transport the technician to various coastal navigation facilities around the BurinPeninsula. On the morning of the accident, the pilot filed a visual flight rules (VFR) flight plan in St.John's for the day's flying in the BurinPeninsula area. The final destination on the flight plan was Marystown with an expected arrival time of 1500. Flight following for CG352 was through the CCG's Marine Communications and Traffic Services (MCTS). CG352 departed St.John's at 1029 with one passenger. After take-off, the pilot contacted MCTS advising of his departure and intentions. The first leg was to the Winterland Airport, located 7.5nm west of Marystown, for a fuel stop. The flight was uneventful until about 4nm east of Marystown, where the helicopter encountered a heavy snow squall. Because of the reduced flight visibility, the pilot descended to a lower altitude, reduced speed, and followed the northern shore of MortierBay towards Marystown. Because of the weather conditions, the helicopter landed at the Marystown shipyard at about 1148. The pilot radioed the MCTS advising of the unscheduled landing, and the helicopter was shut down. About 30minutes after landing, the weather improved significantly. The pilot and passenger cleared the helicopter of snow, and continued the flight to the WinterlandAirport, where the helicopter was fuelled. Another snow squall at the WinterlandAirport delayed departure for about 30minutes. When the weather cleared, the helicopter was flown to a rendezvous point at PointMay on the southern tip of the BurinPeninsula. A change of keepers at the GreenIsland lighthouse was completed at about 1400. The pilot radioed the MCTS that he was en route to the WinterlandAirport for fuel, and would then proceed to the staging area in Marystown to meet with the CCG technician. The helicopter was refuelled once more at the WinterlandAirport, and then was flown to the Marystown staging area, landing at about 1452. The pilot advised MCTS of the helicopter's movements during the flight. Photo1. Go By Point looking north After loading the helicopter, the pilot and his passenger flew to a helicopter landing pad at GoByPoint, at the entrance to MortierBay. This was a very short flight because GoByPoint is only 3nm southeast of Marystown (see AppendixA). The GoByPoint landing pad is situated on steep and rocky terrain next to a marine navigation light (seePhoto1). At 1517, the pilot reported to the MCTS that he had landed at GoByPoint and anticipated one hour of work. He then shut down the helicopter, and he and the passenger disembarked. While at Go By Point, the pilot took several photographs. In two photographs, he captured the navy vessel HerMajesty's Canadian Ship (HMCS) GooseBay approaching the entrance to MortierBay. The photographs showed sunny conditions, and the sky to the west, south, and east was clear with unrestricted visibility. The last photograph was of HMCS GooseBay adjacent to GoByPoint as it sailed north into MortierBay for Marystown. The ship was at this position at 1526. At 1530, security cameras, located at CowHead, 3.5nm north of GoByPoint on the northern shore of MortierBay, showed a heavy snow shower underway. The security cameras recorded uninterrupted snowfall from 1530 until 1630, with intermittent periods of heavy snow and reduced visibility. HMCSGooseBay encountered worsening weather during its transit into Marystown: from sunny and clear at GoByPoint to overcast with visibility reduced at times to 100yards in snow squalls. There were no radio communications from the pilot to MCTS after departure from the GoByPoint pad. Therefore, the exact time of the helicopter's departure from GoByPoint is not known. After departing GoByPoint, the helicopter circled small DuckIsland, which is just east of GoByPoint, at a low altitude. Work was planned the following day at DuckIsland, and the pilot and passenger were likely assessing the landing point at the island. The helicopter then proceeded north over the eastern shore of the entrance to MortierBay following the shoreline around the north side of MortierBay. At that time, it was not yet dark, and light snow was falling at GoByPoint. The helicopter continued north along the eastern shore of the channel entrance, crossing SpanishRoom Point about 1nm south of CowHead. The helicopter then proceeded west towards Marystown along the northern shore of MortierBay. When last observed, the helicopter was about 1nm east of Marystown, flying slowly at low altitude, in heavy snow and in near dark conditions. According to Transport Canada's operations manual for CCG helicopters (OM), flight operations in uncontrolled airspace are not authorized when the reported visibility is less than one (1) mile. Visual reference to the surface must be maintained at all times. If visibility is deteriorating during a flight the PIC [pilot-in-command] must decide to slow down, land and/or reverse course early enough to avoid losing visual reference. As the visibility deteriorates, pilot workload rises appreciably as it becomes necessary to maintain constant attention to outside visual cues. The passenger in the helicopter was carrying a cellular telephone, which was continuously on the cellular network on the day of the accident until 1628. This time is consistent with estimates of the time of the accident. At 1800, one hour after the expected time of arrival as indicated on the flight plan, and in accordance with normal procedures, the helicopter was reported overdue to the Joint Rescue Coordination Centre (JRCC) by the HalifaxFIC. A search was then initiated using ground, sea, and air resources. The bodies of the pilot and passenger were located and recovered from the water near Gould'sCove later that night. Wreckage and Impact Information The helicopter was equipped with an underwater locator beacon (ULB). The ULB is designed to activate upon immersion and to transmit an acoustic signal at 37.5kilohertz(kHz). This signal propagates well in water and is normally easily detected using portable hydrophone detection equipment. An intensive hydrophone search for the beacon was commenced on 09December2005, but the beacon signal could not be detected. Test beacons were then lowered to the bottom. These were easily detectable at distances of over 1sm. A malfunction of the helicopter's ULB was suspected. The area was then searched using side-scanning sonar with remotely operated vehicles investigating the sonar contacts. The helicopter was located on 17December2005, 1000feet northeast of BigHead, in about 100feet of water. All major components were accounted for and were near the main fuselage. The close distribution of wreckage items on the sea bottom was consistent with a helicopter that was intact when it struck the water. The helicopter was recovered on 18December2005 and was shipped to TSB facilities in Dartmouth, Nova Scotia, for further examination. Examination of the helicopter did not reveal any pre-existing mechanical abnormalities that could have contributed to the occurrence. Impact marks showed that there were two distinct collisions with the water. The first impact was tail low, in forward flight. This impact tore open the cloth covering of the pop-out floats, and removed the spoiler from the lower belly of the helicopter. The tail boom, including the tail rotor, would have been immersed in the water. The tail boom was torn from the helicopter fuselage in an upward direction towards the right. The impact also deformed the engine and transmission deck, causing the failure of both engine drive shafts from the main transmission. The helicopter then skipped off the water, and rotated because of the loss of tail rotor drive. The second impact was rearward. Both of the front seats were found with their seat backs leaning markedly backwards. Also, the liferaft mount, which is located between the front seats, failed at the forward attachment bolts, pinning the liferaft firmly against the centre rear passenger seat. Both of the helicopter's engines were operating at the time of impact, with the damage to both engine input shafts suggesting that the engines were delivering significant power. The helicopter is certified for flight in falling or recirculating snow provided the particle separator, engine anti-ice and continuous ignition switches are selected to ON. These switches were found in the OFF position. Personnel Information The pilot held a valid medical certificate and commercial helicopter licence. He was not instrument rated and did not have a night endorsement. The pilot had been flying helicopters for the CCG in Newfoundland for 27years. He was highly regarded as a competent pilot and had accumulated over 20000helicopter flying hours. In the previous 30days, he had flown 21hours, and in the previous 90days, he had flown 42hours, all on the MBBBO105. The pilot had received recurrent ground and flight training in November2005 and had successfully completed an annual pilot proficiency check (PPC) in November2005. He also received pilot decision making (PDM) and human factors training in November2005. Helicopter Information The accident helicopter was manufactured in 1983. It was maintained by TCASD personnel and flown by TCASD pilots. A review of the documentation indicated that the helicopter had been maintained in accordance with existing regulations and approved procedures. All modifications, mandatory airworthiness directives, and required maintenance had been completed. The helicopter had flown about 6530 total hours, including 1438hours since the completion of the last scheduled major airframe inspection (OPS4) on 28December2000. It had undergone a 100-hour inspection on 14October2005, about 42flight hours before the accident flight. The helicopter's technical records did not indicate any outstanding or recurring maintenance items. The helicopter's weight and centre of gravity were within the prescribed limits. The helicopter was not equipped with a flight data recorder or a cockpit voice recorder, nor were they required by regulation. The helicopter was equipped with emergency flotation devices, commonly referred to as pop-out floats. The floats are meant to increase the survivability of a ditching3 by slowing the rate at which a helicopter will sink. If the helicopter is to be flown over water at airspeeds below 60knots, the pilot should normally arm the pop-out floats. The float arm switch was found in the OFFposition. Survival Aspects In a water impact or a capsizing event, the occupants of a helicopter face serious and immediate survival challenges. The need to escape from a capsized helicopter is immediate, but egress can be difficult due to injuries, disorientation, cold water shock, and/or the inability to breath. Commercial offshore helicopter operators and the petroleum industry have recognized the importance of helicopter egress and survival training. In these types of operations, egress and water survival training is mandatory for all crew and passengers. None of those who flew on CG352 on the day of the accident had received helicopter emergency egress/water survival training. Regulations do not require this training, and it was not required by the operator. However, training was offered annually to ASD employees who wished to receive it. Survival Equipment Installations In addition to the emergency pop-out floats that are used during a ditching, CCG helicopters carry a liferaft and, in accordance with applicable Canadian Aviation Regulations (CARs), two emergency locator transmitters (ELTs): a type AF4 and a type W.5 With the exception of the type AFELT, all of this equipment requires manual deployment and activation. In this accident, both ELTs and the liferaft remained inside the helicopter and sank with it. Consequently, no emergency signal was detected by search and rescue (SAR) equipment, and no liferaft was available for the survivors. When the helicopter was recovered, both ELTs and the liferaft were found mounted in their brackets. Had there been an ELT signal from the helicopter, naval and CCG rescue vessels were immediately available to respond. Various survival equipment installation options exist that can improve the probability of surviving a capsized helicopter. Automatically deployed pop-out floats, liferafts, and floating crash position indicators are available. Man-portable backpacks exist, which contain a single-place raft and survival equipment. There are small emergency position indicating radio beacons (EPIRBs) capable of notifying SAR, which can be carried on a lifejacket. The United Kingdom Civil Aviation Authority has recently completed a report on helicopter ditching and crashworthiness.6 The report, which summarizes the results of research undertaken over about 12years, provides further information towards improving the survivability and the safety of offshore operations. Lifejackets and Immersion Suits The OM requires wearing lifejackets on any flight over water. The pilot wore a lifejacket; the passenger did not, although there were several available inside the helicopter. The water surface temperature was 6C. Neither the pilot nor the passenger wore a helicopter passenger transportation (immersion) suit. The pilot was wearing his CCG flight uniform, which had several layers. The passenger was wearing a one-piece insulated coverall. During the investigation of a previous fatal CCG helicopter crash (TSB report A00A0076), the TSB identified regulatory shortcomings regarding requirements for survival equipment and cold water immersion suits during over-water flights. The requirements are based primarily upon the flight time or distance from shore. The CARs requirement to carry a liferaft7 states: No person shall operate over water a multi-engined helicopter that is able to maintain flight with any engine failed at more than 50nautical miles, or the distance that can be covered in 30minutes of flight at the cruising speed filed in the flight plan or flight itinerary, whichever distance is the lesser, from a suitable emergency landing site unless life rafts are carried on board and are sufficient in total rated capacity to accommodate all of the persons on board. The accident helicopter carried a liferaft even though there was no regulatory requirement to do so. The CARs requirement for immersion suit use8 specifies that, where a helicopter is required to carry liferafts, it shall not be operated over water having a temperature of less than 10C unless a helicopter passenger transportation suit is provided for each person on board. Because there was no regulatory requirement for the accident helicopter to carry a liferaft, there was no regulatory requirement to wear an immersion suit. On 26 February 2001, the TSB sent a Safety Advisory (A010009-1) to TCCivil Aviation suggesting that TCCivil Aviation consider revising the criteria for survival equipment carriage and use on over-water flights so that the criteria would be more relevant than time and distance requirements alone. On 02April2001, TCCivil Aviation responded to the Safety Advisory. It agreed that changes to the regulations may be warranted, and stated that it would convene a working group to examine the issue of offshore operations and make recommendations. The OM requirements for immersion suit use are more stringent than those found in CARs. The OM states that immersion suits are mandatory in multi-engine helicopters when they are flown over water more than 15nm from ship, shore, or continuous ice capable of supporting the helicopter. Under these criteria, immersion suit use was not mandatory for the coastal flights flown on the day of the accident because, although much of the flying done by CG352 was over water, none of it was beyond 15nm from shore. The use of immersion suits is encouraged in the OM. However, in practice, the suits are not normally worn on over-water CCG flights unless the mandatory conditions apply. The accident pilot had been provided with two immersion suits: an older wet suit and a newer dry suit. Neither suit was used. Although passenger transportation suits were available in St.John's, none of the passengers flown that day were aware of or offered passenger transportation suits. Direct-to-Airframe Helmet Connection After the accident, an examination of the pilot's aviation helmet found that the end fitting of the communication cord was fractured at the point where it attached to the helicopter. The communication cords for front seat occupants connect to receptacles located on the overhead centre console. When the helicopter was recovered, the metal pins from the end fitting were still inside the receptacle. Metal remnants from the connection show that the cord was being pulled sideways, towards the pilot's door, when the fracture occurred. A downward pull is required to release the connection. A break test of a similar fitting required a 70-pound pull before the cord failed. After water impact or a helicopter capsizing, a quick unimpeded egress through any available exit is vital to survival. An attached communication cord that will not release cleanly may impede this egress. In the past, CCGBO105 helicopters had been fitted with an intermediate pig-tail communication cord for helmet connections. Instead of plugging the helmet cord into the helicopter's receptacle, the helmet cord was instead plugged into this intermediate cord. The helmet connection plug can release cleanly from the intermediate pig-tail cord receptacle as it is pulled in the direction of travel during egress. Liferaft Mounting Bracket Failure When the helicopter struck the water, the secondary and most substantial impact was rearward. During the rearward impact, the liferaft mounting bracket, located between and just behind the two cockpit seats, failed. When the helicopter was recovered, the liferaft was found pinned against the centre rear passenger seat. The liferaft mounting bracket forward attachment bolts had pulled through the mount frame. When the forward right attachment failed, the base of the mounting bracket rode up over the seat belt attachment bolts, and the bracket was effectively jammed in this aft position. The liferaft mount installation was in accordance with a Limited Supplemental Type Certificate (LSTCO-LSH94-2029/D) issued to TCASD on 22December1994. At the time of approval of the LSTC, the mounting frame was analyzed following criteria in Section27.561 of the UnitedStates Federal Aviation Regulations (FARs) for Emergency Conditions (1-1-88 Edition). This edition of the FARs did not require an analysis of rearward or sideward impact forces. Therefore, no analysis was done for these conditions. The current edition of Section27.561 of the FARs requires that sideward and rearward impact forces be considered. Underwater Locator Beacon Failure The ULB from the accident helicopter (modelDK120, serial numberDT1218) was examined by the TSB Engineering Laboratory. When the ULB was placed in room temperature water, it transmitted normally. However, when the water temperature was lowered to near freezing, the signal quickly dropped off and was no longer detectable using the hydrophone detection gear. Other ULBs from TC ASD helicopters were also examined, and one (serial numberDT1226, which was removed from ASD maintenance stores) was found to have the same signal failure at low temperatures. It was found that the signal failure was due to the delamination of a metal coating on a ceramic resonating ring. The delamination occurred immediately beneath a solder joint on the outer portion of the ring. The combined effect of the delamination and the cold temperatures was to remove the electrical connection to the ring, shifting the beacon signal frequency from 37kHz to 166kHz, which is beyond the detectable range of the locating equipment. Transport Canada Aircraft Services Directorate Operational Issues TC ASD operates a mixed fleet of helicopters and fixed-wing aircraft. TCASD supports helicopter and fixed-wing flight operations for the Department of Fisheries and Oceans Canadian Coast Guard. In addition, TCASD supports flight operations for other government agencies. A review of previous TCASD accidents and incidents has shown a recurrence of previously identified safety issues. On 10 May 2000, a CCG Bell212 helicopter crashed into the water near CabotIsland, Newfoundland and Labrador, killing the pilot (TSB report A00A0076). The TSB investigation found that the pilot was neither equipped to survive either a less severe accident or a controlled ditching into the frigid water, wearing an immersion suit or a lifejacket, nor wearing the supplied shoulder harness, as required by CARs. The TSB sent a Safety Information Letter (A000048-1) and two Safety Advisories (A010009-1 and A010006-1) to TCCivil Aviation concerning these issues, and identified these issues in its final report. On 07 May 2005, a CCG BO105 helicopter crashed during slinging operations near BellaBella, British Columbia (TSB reportA05P0103). The pilot was not wearing the supplied shoulder harness. On 31May2005, the TSB sent a Safety Information Letter (A050014-1) to TCCivil Aviation and TCASD, outlining that, despite the CARs requirement and a previous Safety Advisory (A010006-1) in 2001, helicopter slinging operations without upper-body restraint were still occurring at TCASD. Also related to the BellaBella accident, the TSB sent a Safety Advisory (A050015-1) to TCCivil Aviation and TCASD pointing out that the currently installed liferaft mount fixtures in their BO105helicopters presented a head strike hazard during an accident. Although the risk of head injury was mitigated somewhat for the pilots because they wear helmets as required by the OM, any passenger seated in the left front seat would be exposed to this risk because the OM does not require helmet use by passengers. The Safety Advisory suggested that TCASD may wish to modify the fixtures, or limit use of the front seats to persons wearing protective head gear. On 20September2005, TCASD responded to the Safety Advisory, stating that it would undertake a complete review of the applicable LSTC. TCASD also indicated that it would consider adding padding to the fixture, and that it would review the requirement for helmets to be worn in helicopters with this installation. The pilot of CG352 was wearing an aviation helmet. The passenger in CG352 was seated in the front seat and was not wearing protective head gear. As of December2005, the suggested modifications had not been made to the liferaft mounts, protective padding had not been added, and passengers were still being flown in the front seat without helmets. Line BO105pilots contacted by TSB investigators were unaware of the head strike hazard. TCASD management had not formally warned line pilots of the head strike hazard, or required that front seat passengers wear protective head gear.